extracorporeal ultraﬁltration for fluidoverloadinheartfailure · dan negoianu, md,q margaret m....

Listen to this manuscript’s

audio summary by

JACC Editor-in-Chief

Dr. Valentin Fuster.

J O U R N A L O F T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y VO L . 6 9 , N O . 1 9 , 2 0 1 7

ª 2 0 1 7 T H E A U T H O R S . P U B L I S H E D B Y E L S E V I E R O N B E H A L F O F T H E AM E R I C A N

C O L L E G E O F C A R D I O L O G Y F O U N DA T I O N . T H I S I S A N O P E N A C C E S S A R T I C L E U N D E R

T H E C C B Y - N C - N D L I C E N S E ( h t t p : / / c r e a t i v e c o mm o n s . o r g / l i c e n s e s / b y - n c - n d / 4 . 0 / ) .

I S S N 0 7 3 5 - 1 0 9 7

h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j a c c . 2 0 1 7 . 0 3 . 5 2 8

THE PRESENT AND FUTURE

STATE-OF-THE-ART REVIEW

Extracorporeal Ultrafiltration forFluid Overload in Heart FailureCurrent Status and Prospects for Further Research

Maria Rosa Costanzo, MD,a Claudio Ronco, MD,b,c William T. Abraham, MD,d Piergiuseppe Agostoni, MD,e,f

Jonathan Barasch, MD, PHD,g Gregg C. Fonarow, MD,h Stephen S. Gottlieb, MD,i,j Brian E. Jaski, MD,k,l

Amir Kazory, MD,m Allison P. Levin, BA,n Howard R. Levin, MD,o Giancarlo Marenzi, MD,e Wilfried Mullens, MD,p

Dan Negoianu, MD,q Margaret M. Redfield, MD,r W.H. Wilson Tang, MD,s Jeffrey M. Testani, MD, MTR,t

Adriaan A. Voors, MD, PHDu

ABSTRACT

Fro

Ho

Me

ScigD

Ah

Me

mo

Ne

De

LL

Me

MesD

MeuU

Co

AV

Ba

Me

co

Ga

More than 1 million heart failure hospitalizations occur annually, and congestion is the predominant cause.

Rehospitalizations for recurrent congestion portend poor outcomes independently of age and renal function.

Persistent congestion trumps serum creatinine increases in predicting adverse heart failure outcomes. No

decongestive pharmacological therapy has reduced these harmful consequences. Simplified ultrafiltration devices

permit fluid removal in lower-acuity hospital settings, but with conflicting results regarding safety and efficacy.

Ultrafiltration performed at fixed rates after onset of therapy-induced increased serum creatinine was not superior

to standard care and resulted in more complications. In contrast, compared with diuretic agents, some data suggest

that adjustment of ultrafiltration rates to patients’ vital signs and renal function may be associated with more

effective decongestion and fewer heart failure events. Essential aspects of ultrafiltration remain poorly defined.

Further research is urgently needed, given the burden of congestion and data suggesting sustained benefits of early

and adjustable ultrafiltration. (J Am Coll Cardiol 2017;69:2428–45) © 2017 The Authors. Published by Elsevier on

behalf of the American College of Cardiology Foundation. This is an open access article under the CC BY-NC-ND

license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

m the aAdvocate Heart Institute, Naperville, Illinois; bDepartment of Nephrology, Dialysis and Transplantation, San Bortolo

spital, Vicenza, Italy; cInternational Renal Research Institute of Vicenza (IRRIV), Vicenza, Italy; dDivision of Cardiovascular

dicine, The Ohio State University, Columbus, Ohio; eCentro Cardiologico Monzino, Istituto di Ricovero e Cura a Carattere

entifico (IRCCS), Milan, Italy; fDepartment of Clinical Sciences and Community Health, University of Milan, Milan, Italy;

epartment of Medicine, Columbia University College of Physicians and Surgeons, New York, New York; hDivision of Cardiology,

manson University of California at Los Angeles Cardiomyopathy Center, Los Angeles, California; iDivision of Cardiovascular

dicine, University of Maryland School of Medicine, Baltimore, Maryland; jBaltimore Veterans’ Affairs Medical Center, Balti-

re, Maryland; kSharp Healthcare, San Diego, California; lSan Diego Cardiac Center, San Diego, California; mDivision of

phrology, Hypertension, and Renal Transplantation, University of Florida, Gainesville, Florida; nDivision of Cardiology,

partment of Medicine, Columbia University Medical Center, New York Presbyterian Hospital, New York, New York; oCoridea,

C, New York, New York; pDepartment of Cardiology, Ziekenhuis Oost Limburg, Genk-Biomedical Research Institute, Faculty of

dicine and Life Sciences, Hasselt University, Diepenbeek, Belgium; qDivision of Nephrology, University of Pennsylvania

dical Center, Philadelphia, Pennsylvania; rDepartment of Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota;

epartment of Cardiovascular Medicine, Heart and Vascular Institute, Cleveland Clinic, Cleveland, Ohio; tDepartment of Internal

dicine, Program of Applied Translational Research, Yale University School of Medicine, New Haven, Connecticut; and the

niversity of Groningen, University Medical Center Groningen, Department of Cardiology, Groningen, the Netherlands. Dr.

stanzo served as principal investigator for AVOID-HF trial; has received research support through her institution for the

OID-HF trial; consultant for Axon Therapies. Columbia University is the assignee for biomarker patents developed by Dr.

rasch. Dr. Fonarow has received funding from National Institutes of Health; and is consultant for Amgen, Janssen,

dtronic, Novartis, and St. Jude Medical. Dr. Gottlieb has received research grants from Amgen and Novartis; and is

nsultant for Bristol-Myers Squibb. Dr. Levin holds equity in Coridea and Axon Therapies. Dr. Negoianu is a speaker for

mbro Inc./Baxter and Fresenius; and was a member of the Steering Committee for AVOID-HF trial. Dr. Voors has received

http://creativecommons.org/licenses/by-nc-nd/4.0/

https://s3.amazonaws.com/ADFJACC/JACC6919/JACC6919_fustersummary_07




http://crossmark.crossref.org/dialog/?doi=10.1016/j.jacc.2017.03.528&domain=pdf

http://creativecommons.org/licenses/by-nc-nd/4.0/

http://dx.doi.org/10.1016/j.jacc.2017.03.528

J A C C V O L . 6 9 , N O . 1 9 , 2 0 1 7 Costanzo et al.M A Y 1 6 , 2 0 1 7 : 2 4 2 8 – 4 5 Ultrafiltration for Fluid Overload in Heart Failure

2429

AB BR E V I A T I O N S

AND ACRONYM S

NGAL = neutrophil gelatinase-

associated lipocalin

UF = ultrafiltration

A nnual hospitalizations for heart failureexceed 1 million in both the United Statesand Europe, and more than 90% are due to

symptoms and signs of fluid overload. In addition,up to 1 in 4 patients (24%) are readmitted within 30days, and 1 in 2 patients (50%) are readmitted within6 months (1,2). Recurrent fluid overload in heart fail-ure has uniformly been associated with worse out-comes independently of age and renal function (3).

PATHOPHYSIOLOGICAL CONSEQUENCES OF

FLUID OVERLOAD

Compared with normal subjects, asymptomaticpatients with heart failure have decreased sodiumexcretion in response to volume expansion (4).Abnormal fluid handling leads to physiologicalabnormalities in multiple organ systems. Increasedmyocardial water can lead to ischemia and decreasedcontractility in animals and humans (5–8). Derangedhemodynamics, neurohormonal activation, excessivetubular sodium reabsorption, inflammation, oxida-tive stress, and nephrotoxic medications are impor-tant drivers of harmful cardiorenal interactions inpatients with heart failure (8–10).

Elevation of central venous pressure is rapidlytransmitted to the renal veins, causing increasedinterstitial and tubular hydrostatic pressure, whichdecreases net glomerular filtration (9,11,12). Anincreased central venous pressure is independentlyassociated with renal dysfunction and unfavorableoutcomes in both acute and chronic heart failure(13,14). Venous congestion itself can produce endo-thelial activation, up-regulation of inflammatorycytokines, hepatic dysfunction, and intestinal villiischemia (15). Bacterial endotoxins can then enter thecirculation, magnifying the inflammatory milieucreated by venous congestion and neurohormonalactivity (8).

Three recent studies suggest that failure toadequately reduce fluid excess in patients withacutely decompensated heart failure trumpsincreases in serum creatinine in predicting pooroutcomes (16). Thus, the foremost goal in managingacutely decompensated heart failure is to effectivelyresolve fluid overload (16). Therefore, if a decrease inintravascular volume by fluid removal causes smalltransient increases in serum creatinine, effective

research grants and consultancy fees from AstraZeneca, Bayer, Bristol-Myer

GlaxoSmithKline, Merck/MSD, Novartis, Servier, Sphingotec, Stealth Peptid

Coridea, LLC. All other authors have reported that they have no relationshi

Manuscript received March 3, 2017; accepted March 7, 2017.

decongestion may still be essential to protectthe kidney in the long term (16,17). With-drawal of diuretic agents in 30 euvolemicpatients with heart failure resulted in in-creases in urinary levels of kidney injurymolecule-1, which returned to baseline with
resumption of diuretic agents. Thus, in heart failure,even subclinical fluid overload can be associated withbiological evidence of tubular dysfunction (18). Anunresolved challenge is the ability to discern whetherincrease in serum creatinine during fluid removal isdriven primarily by hemodynamic decreases inglomerular filtration rate or by development of acutetubular damage, which can progress to chronic kidneydisease (19).
UNRESPONSIVENESS TO DIURETIC AGENTS

IN HEART FAILURE

Diuretic agents remain the cornerstone of therapy forfluid overload. Although effective early in heartfailure, diuretic agents become increasingly ineffec-tive with disease progression due to the developmentof unresponsiveness in a significant subset of patients(20). Excellent reviews describe the mechanismsleading to decreased diuretic agent responsiveness(21). In patients with heart failure, impaired absorp-tion, decreased renal blood flow, azotemia, and pro-teinuria all result in reduced levels of active diureticagents in the tubular lumen (21). Recently proposeddefinitions of diuretic resistance include persistentcongestion, despite adequate and escalating doses ofdiuretic agents equivalent to $80 mg/day furose-mide; the amount of sodium excretion as a percent-age of filtered load below 0.2%; and failure to excreteat least 90 mmol of sodium within 72 h of a 160-mgtwice-daily dose of furosemide. Metrics for diureticagent response have also been proposed, includingweight loss per 40 mg of furosemide or equivalent;net fluid loss per milligram of loop diuretic agent; andnatriuretic response to furosemide as urinary sodium-to-urinary furosemide ratio (21).

The clinical hallmarks of diuretic agent resistanceare insufficient symptom relief, higher risk ofin-hospital worsening of heart failure, increasedmortality after discharge, and a 3-fold increase inrehospitalization rates (21,22). Among more than50,000 patients enrolled in the ADHERE (Acute

s Squibb, Boehringer Ingelheim, Cardio3Biosciences,

es, Trevena, and Vifor. Dr. Stough was funded by

ps relevant to the contents of this paper to disclose.

Costanzo et al. J A C C V O L . 6 9 , N O . 1 9 , 2 0 1 7

Ultrafiltration for Fluid Overload in Heart Failure M A Y 1 6 , 2 0 1 7 : 2 4 2 8 – 4 5

2430

Decompensated Heart Failure National Registry)study, only 33% lost $2.27 kg (5 lbs), and 16% gainedweight during hospitalization. Nearly one-half ofhospitalized patients with heart failure are dis-charged with residual fluid excess after receivingconventional diuretic therapies (23). Regardless ofdiuretic strategy, 42% of acutely decompensatedheart failure subjects in the DOSE (Diuretic Optimi-zation Strategies Evaluation) trial reached the com-posite endpoint of death, rehospitalization, oremergency department visit at 60 days (24). Vaso-pressin and adenosine-A1 receptor antagonists,exogenous natriuretic peptides, and low-dose dopa-mine, studied as either a complement or a replace-ment for conventional diuretic therapies, candecrease fluid overload in the short term but havefailed to improve long-term outcomes (25–27).

Therefore, there is a clear, unmet clinical need foralternative methods of fluid removal with superiorefficacy in patients with heart failure. One therapythat might prove successful is extracorporeal ultra-filtration (UF) (28,29). Greater access to UF has beenfacilitated by the development of simplified devicesthat do not require specialized technicians or acutecare settings (Online Table 1) (30).

Over the past 20 years, several small studies haveattempted to define the physiological rationale forthe clinical benefits of mechanical fluid removal byUF in heart failure (31–35). However, concerns arosefrom reports of treatment-related adverse events (32).Thus, the principal aims of this paper were to reviewthe available data for the use of UF in patients withheart failure, describe the knowledge gaps in thisarea, and outline potential future studies to answerunresolved questions.

PROCESS OF FLUID REMOVAL BY UF,

HEMOFILTERS, PUMPS, AND

VASCULAR ACCESS

Ultrafiltration consists of the production of plasmawater from whole blood across a semipermeablemembrane (hemofilter) in response to a trans-membrane pressure gradient (29). The newer,simplified UF devices afford the advantages of smallsize, portability, low blood flow rates, and anextracorporeal blood volume below 50 ml. Thesedevices provide a wide range of UF rates (0 to 500ml/h) and do not mandate admission to intensivecare units or cannulation of a central vein. Thecharacteristics of 2 of these devices are shown inFigure 1. Additional details for hemofilters, pumps,and vascular access for UF can be found in OnlineAppendix in Section 1.0.

DIFFERENTIAL CHARACTERISTICS OF

DIURETIC AGENT AND UF-BASED

FLUID REMOVAL

Loop diuretic agents selectively block the Naþ/Kþ/2Cl� cotransporter in the luminal membrane of themedullary thick ascending loop of Henle. Edema inpatients with heart failure is isotonic, and thereforenormonatremic edematous patients have signifi-cantly increased total body sodium. Because loopdiuretic agents inhibit sodium reabsorption at a sitein the kidney also critical for water reabsorption, theygenerally result in greater loss of water than sodiumand therefore generate hypotonic urine (29).

In contrast, because the ultrafiltrate is almostiso-osmotic and isonatremic compared to plasma,approximately 134 to 138 mmol of sodium areremoved with each liter of ultrafiltrate (29). Thus, forany amount of fluid withdrawn, more sodium is likelyto be removed with UF than with diuretic agents (29).Conversely, with these drugs, changes in intravas-cular volume are unpredictable. Furthermore, loopdiuretic agents inhibit sodium chloride uptake in themacula densa, an event that, coupled withaugmented release of prostacyclin, enhances renalsecretion of renin (9,29). These effects augmentneurohormonal activation, which ultimately reducesdiuretic agents’ effectiveness (9). As opposed to thedirect effect of loop diuretic agents on the maculadensa, with UF, neurohormonal activation shouldoccur only if the fluid removal rate causes intravas-cular volume depletion by exceeding the plasmarefilling rate (Table 1) (35). This measure of plasmawater transport from the interstitium into thevasculature during fluid removal varies betweenpatients depending upon serum albumin concentra-tion (i.e., serum oncotic pressure) and capillarypermeability.

CLINICAL RESEARCH PRECEDING

CONTROLLED TRIALS OF UF IN

HEART FAILURE

Studies of extracorporeal fluid removal conductedbefore the introduction of contemporary UF devicesare summarized in Online Table 2. The key lessonsfrom these early investigations are that UF canrestore diuretic agent responsiveness, but overlyaggressive fluid removal can convert nonoliguricrenal dysfunction into oliguric failure and dialysisdependence (34,35). A more detailed description ofthe early studies that specifically evaluated themechanisms of action of extracorporeal fluid removalcan be found in Online Appendix in Section 2.0.






FIGURE 1 UF Circuit

(A) The console controls blood removal rates and extracts ultrafiltrate at a maximum rate set by the clinician. Blood is withdrawn from a vein through the

withdrawal catheter (red) connected by tubing to the blood pump. Blood passes through the withdrawal pressure sensor before entering the blood pump

tubing loop. After exiting the blood pump, blood passes through the air detector and enters the hemofilter (made of a bundle of hollow fibers) through a

port on the bottom, exits through the port at the top of the filter, and passes through the infusion pressure sensor before returning to the patient (blue).

The ultrafiltrate passes sequentially through the ultrafiltrate’s pressure sensor, the pump, and the collecting bag suspended from the weight scale. A

hematocrit sensor is located on the withdrawal line. (B) This UF system requires only a single-lumen, multihole, small (18-gauge) cannula inserted in a

peripheral vein of the arm. A syringe pump drives the blood inside the extracorporeal circuit, which includes 2 check valves that allow the blood to move

from the vein to the filter, and then returns it to the same vein through alternate flows that can be independent. The priming volume of 50 ml and the

reduced contact surface between blood and tubing set ensure minimal blood loss if circuit clots and for reduced heparin requirements. BD ¼ blood

detector; BLD ¼ blood leak detector; HTC ¼ hematocrit sensor; UF ¼ ultrafiltration.


2431

TABLE 1 Comparative Characteristics of Loop Diuretic Agents

and Isolated UF

Loop Diuretic Agents Isolated UF

Direct neurohormonal activation No direct neurohormonal activation

Elimination of hypotonic urine Removal of isotonic plasma water

Unpredictable elimination ofsodium and water

Precise control of rate and amountof fluid removal

Development of diuretic agentresistance with prolongedadministration

Restoration of diuretic agentresponsiveness

Risk of hypokalemia andhypomagnesemia

No effect on plasma concentrationof potassium and magnesium

Peripheral venous access Peripheral or central venous catheter

No need for anticoagulation Need for anticoagulation

No extracorporeal circuit Need for extracorporeal circuit

UF ¼ ultrafiltration.



2432

PILOT STUDIES WITH

CONTEMPORARY UF SYSTEMS

The SAFE (Simple Access Fluid Extraction) trialshowed that, in 21 congested patients with heartfailure, removal of an average of 2,600 ml of ultra-filtrate during one 8-h treatment session reducedweight by an average of 3 kg without changes in heartrate, blood pressure, serum creatinine concentration,and electrolytes or the occurrence of major adverseevents (30). Results of 2 additional pilot studies withthis system are summarized in Online Table 3.

The findings of both studies provided valuable in-formation. First, the clinical benefits of UF can persistbeyond the index heart failure hospitalization up to90 days. Second, UF is unlikely to improve outcomesof patients with end-stage heart failure and should beused very cautiously in this setting. Third, althoughpotentially effective in reducing central venouspressure, aggressive fluid removal in preload-dependent patients with heart failure, who have lowforward flow as the predominant mechanism for areduction in glomerular filtration rate, can rapidlydecrease renal perfusion pressure and cause oliguricfailure, leading to dialysis dependence (36–38). Highpre-treatment intravenous loop diuretic agent dosesmay increase the risk of tubular injury when addi-tional fluid is removed by UF (39).

RANDOMIZED CONTROLLED TRIALS

The randomized controlled trials of UF are summa-rized in Table 2 and Online Table 4. The RAPID-CHF(Relief of Acutely Fluid-Overloaded Patients withDecompensated Congestive Heart Failure) trial wasthe first randomized study of UF to use the AquadexSystem 100 device (Sunshine Heart, Minneapolis,

Minnesota) (40). Although this small study did notevaluate patients’ outcomes beyond 48 h, itconfirmed that effective fluid removal and clinicalimprovement may occur with UF (41).

The physiological fact that refill of the intravas-cular space from the edematous interstitium de-creases as fluid is removed led to the hypothesis thatinitiation of UF before the plasma refilling rate isdecreased by previous diuretic agent-based therapiesmight produce greater benefit than intravenous loopdiuretic agents in unequivocally congested patientswith heart failure. Hence, in the UNLOAD (Ultrafil-tration Versus Intravenous Diuretics DecompensatedHeart Failure) trial, randomization had to occurwithin 24 h of hospitalization, and a maximum of 2intravenous loop diuretic agent doses were permittedbefore enrollment (33). Compared with standard careresults, the UF group had greater weight loss andsimilar improvement in dyspnea score (the coprimaryendpoints) at 48 h. The percentage of patients withincreases in serum creatinine levels $0.3 mg/dl wasslightly but insignificantly higher in the UF groupthan in the control group at 24 and 48 h (33). AmongUNLOAD patients from a single center, use of iotha-lamate and para-aminohippurate to measureglomerular filtration rate and renal plasma flowshowed that UF and furosemide produce similarchanges in these variables (41). There were nobetween-group differences in duration of the indexhospitalization, a variable that can be influenced byadjustment of oral heart failure therapy beforedischarge, performance of additional diagnostic andtherapeutic procedures, treatment of comorbidities,issues of patients’ placement, and lack of defineddischarge criteria (33,42). In UNLOAD, the 90-dayheart failure events were a pre-specified secondaryendpoint, and the investigators determined whetherthese were related to worsening heart failure or not. Itcannot be said with certainty whether the fewer heartfailure events in 90 days in the UF group comparedwith the standard care group were due to differencesin fluid loss, the nature of the fluid removed, or otherfactors (33). Because UNLOAD did not have an inde-pendent clinical events committee (CEC) to adjudi-cate whether an event was heart failure-related ornot, the possibility of patient or investigator biascannot be excluded. A post hoc analysis of UNLOADcompared the 100 UF patients with 100 usual-carepatients subdivided according to intravenousdiuretic agent strategy (continuous [n ¼ 32] or bolus[n ¼ 68] administration) (43). Despite removal of thesame amount of fluid by UF and diuretic agent infu-sion, 90-day heart failure events were fewer in the UFgroup (p ¼ 0.016) (43). The simultaneous reduction of




2433

total body sodium and excess fluid by UF may bemore effective than removal of hypotonic fluid bydiuretic agents or free water by arginine vasopressinV2 receptor antagonists (25,43,44). It is also possiblethat pre-hospitalization diuretic agent use itselfimpairs the natriuretic response to subsequentintravenous administration of these drugs (21). InUNLOAD, complications related to UF includedclotting of 5 filters, 1 catheter infection, and therequirement for hemodialysis in 1 patient deemedunresponsive to UF (Table 2) (33).

The UNLOAD trial lacked treatment targets, bloodvolume assessments, cost analysis, and adjudicationof events by an independent CEC. Nevertheless,compared with standard care, UF initiated before theadministration of high-dose intravenous diureticagents led to greater fluid loss at 48 h and reduced90-day heart failure events. In the ULTRADISCO(Effects of ULTRAfiltration vs. DIureticS on clinical,biohumoral and hemodynamic variables in patientswith deCOmpensated heart failure) study, at 36 h,compared with the diuretic agent group, the UFpatients had greater reduction in body weight, signsand symptoms of heart failure, aldosterone andN-terminal pro-B-type natriuretic peptide levels, andsystemic vascular resistance, as well as greaterimprovements in objective measures of cardiac per-formance (45). Albeit very small, this study suggeststhat effective fluid removal may be associated withimproved cardiac function (35,45–47).

The CARRESS-HF (Cardiorenal Rescue Study inAcute Decompensated Heart Failure) trial comparedthe effects of UF delivered at a fixed rate of 200 ml/hwith those of stepped pharmacological therapyinclusive of adjustable doses of intravenous loopdiuretic agents, thiazide diuretic agents, vasodilators,and inotropes in acutely decompensated patientswith heart failure who had experienced a pre-randomization increase in serum creatinine(19,32,48). The primary endpoint of CARRESS-HF wasthe bivariate change in serum creatinine and bodyweight from baseline to 96 h after randomization (32).According to the CARRESS-HF design (48), thisprimary endpoint assumes that weight loss is ameasurement of effective fluid removal and that anincrease in serum creatinine represents acute tubularinjury. In CARRESS-HF, both groups lost an equiva-lent amount of weight, but greater increases in serumcreatinine occurred with UF (32). In addition, a higherpercentage of patients in the UF group experiencedserious adverse events (Table 2) (32). However, thefact that 37 patients (39%) in the UF group receivedonly diuretic agents or were given these drugs beforethe assessment of the primary endpoint at 96 h

impairs adjudication of adverse events to one or theother therapy.

Although in heart failure increases in serumcreatinine ($0.3 mg/dl) have been equated to actualrenal tubular damage, which portends adverselong-term prognosis, transient increases in serumcreatinine may simply reflect a hemodynamicallydriven reduction in glomerular filtration rate akin tothat occurring with angiotensin-converting enzymeinhibitors. Furthermore, recent studies suggest thattransient increases in serum creatinine may reflectmore complete decongestion and, instead, forecastimproved post-discharge outcomes (17). Crossoverrates in CARRESS-HF also impaired interpretation ofthe findings of a recent substudy where plasma reninactivity and aldosterone levels were higher in the UFgroup than in the stepped pharmacological therapygroup at 96 h. Because neurohormonal levels werenot measured beyond this time point, it remains un-known whether the observed neurohormonalchanges were transient or sustained (49).

In CARRESS-HF, the rate of fluid removal wasmandated to be 200 ml/h in all UF patients, andadjustments were left to the discretion of theinvestigators, “to address technical problems orclinical care requirements” (32). A UF rate of 200 ml/hmay be excessive for patients with a lower bloodpressure and greater dependence on preload for he-modynamic stability (50,51). Clinical experienceshows that, regardless of the method used, removalof fluid must be tailored to individual patients’ bloodpressure, renal function, urine output, and bodymass. The stepped pharmacological therapy patientscould receive care tailored to their characteristics,including use of vasoactive drugs, which occurred in12% of patients in this arm before 96 h (32). Vasoac-tive agents were prohibited in the UF group, except asrescue therapy. Interpretation of the results of theCARRESS-HF trial is also hampered by the fact thatoverall outcomes were poor, regardless of fluidremoval strategy: only 10% of the patients hadadequate improvement of signs of fluid overload at96 h, and more than 30% died or were readmitted fordecompensated heart failure within 60 days (32,52).

In the CUORE (Continuous Ultrafiltration forCongestive Heart Failure) trial, UF-treated patientshad a lower incidence of heart failure rehospitaliza-tions through 1 year (53) than those undergoingstandard care, despite similar weight loss atdischarge. In CUORE, diuretic agent therapy wascontinued during UF in the belief that this approachmight help restore diuretic agent responsiveness byenhancing urinary sodium excretion (53). In previousstudies, diuretic agent therapy was stopped during

TABLE 2 UF Clinical Trials: Overview of Study Designs and Key Findings

Study Name, PublicationYear (Ref. #) Study Group UF Arm Comparison Arm Primary Efficacy Endpoint

RAPID-HF, 2005 (40) N ¼ 40Hospitalized with HF, 2þ edema

and $1 additional sign ofcongestion

Single, 8-h course, median duration8 h, median volume removed3,213 ml

Standard HF therapies determinedby treating physician

Weight loss 24 h post-consent

UNLOAD, 2007 (33) N ¼ 200Hospitalized with HF, $2 signs

of fluid overload

Aquadex System 100†Mean fluid removal rate 241 ml/h

for 12.3 � 12 h

Standard care: IV diuretic agents.For each 24-h period, at leasttwice the pre-hospitalizationdaily oral dose

Weight loss and dyspneaassessment at 48 h afterrandomization

CARRESS-HF, 2012 (32) N ¼ 188Hospitalized with HF,$2 signs of

congestion, and recent$0.3mg/dl sCr increase

Aquadex System 100† at a fixedrate of 200 ml/h

Median duration 40 h

SPT with intravenous diuretic agentsdosed to maintain urine output3–5 l/day

Bivariate response of change insCr and change in weight96 h after randomization

CUORE, 2014 (53) N ¼ 56NYHA III or IV, LVEF#40%,$4 kg

weight gain from peripheral fluidoverload, over 2 months

Dedyca device‡Mean treatment duration 19 � 90 h;

volume removed 4,254� 4,842 ml

Intravenous diuretic agentsaccording to guidelinerecommendations (standardcare)

HF rehospitalization at 1 yr

AVOID-HF, 2016 (56) N ¼ 224Hospitalized with HF; $2 criteria

for fluid overload; receivingdaily oral loop diuretic agents

AUF with Aquadex FlexFlowSystem§; adjustments perprotocol guidelines on the basisof vital signs and renalfunctionk

Mean fluid removal rate 138 � 47ml/h for 80 � 53 h

ALD with adjustments perprotocol-guidelines on thebasis of vital signs and renalfunction¶

Mean furosemide-equivalent dose271.26 � 263.06 mg for100 � 78 h

Time to first HF event (HFrehospitalization orunscheduled outpatient oremergency treatment withintravenous loop diureticagents or UF) within 90 daysof hospital discharge

ULTRADISCO, 2011 (45) N ¼ 30Hospitalized for HF, $2þ

peripheral edema, $1 othercriteria for volume overload

PRISMA#Median treatment duration 46 h;

cumulative fluid loss 9.7 � 2.9 l

Furosemide continuous infusion,initial dose 250 mg/24 h

Change in hemodynamicsmeasured by PRAM

*Other than primary endpoint. †CHD solutions, Minneapolis, Minnesota. ‡Dellco, Mirandola, Italy. §Baxter International, Deerfield, Illinois. kSee flow chart in Figure 2. ¶See flow chart in Figure 3. #HospalGambro Dasco, Medolla, Italy.

ALD ¼ adjustable loop diuretic agent; AUF ¼ adjustable ultrafiltration; CI ¼ confidence interval; CPO ¼ cardiac power output; dP/dtmax ¼ maximal rate of rise in left ventricular pressure; HF ¼ heart failure;HR ¼ hazard ratio; LVEF ¼ left ventricular ejection fraction; NYHA ¼ New York Heart Association; PRAM ¼ pressure recording analytical method; SAE ¼ serious adverse event; sCr ¼ serum creatinine;SPT ¼ stepped pharmacological therapy; UF ¼ ultrafiltration.

Continued on the next page



2434

extracorporeal fluid removal on the basis of thehypothesis that UF may give patients a “diureticholiday,” during which loop diuretic agent-inducedneurohormonal activation does not occur (33,54,55).The AVOID-HF (Aquapheresis versus IntravenousDiuretics and Hospitalization for Heart Failure) trialtested the hypothesis that patients hospitalized forheart failure who were treated with adjustable UFwould have a longer time to first heart failure eventwithin 90 days than those receiving adjustableintravenous loop diuretic agents (56). The AVOID-HFtrial, designed as a multicenter, 1:1 randomized study

of 810 hospitalized patients with heart failure, wasterminated unilaterally and prematurely by thesponsor (Baxter Healthcare, Deerfield, Illinois) afterenrollment of 224 patients (27.5%). Detailed guide-lines were provided to the investigators as to how toadjust both of the therapies in response to patients’vital signs, renal function, and urine output (Figures 2and 3) (57). Patients in the adjustable UF group had anonstatistically significant trend to longer time tofirst heart failure event than patients in the adjust-able diuretic agents group (62 vs. 34 days, respec-tively; p ¼ 0.106). Although the primary outcome did

TABLE 2 Continued

Primary Endpoint Result Reported Clinical Outcomes* Mortality Adverse Events

Weight loss approximately �6.25 kg (UF)vs. �7 kg (standard care), p ¼ 0.24

Index length of stay: 6 days (UF) vs. 5 days(standard care); p ¼ NS

Volume removal 24 h after consent:4,650 ml (UF) vs. 1,838 ml (standard care),p ¼ 0.001

30 days: 1 (UF) 1 catheter site infection (UF)

Weight loss: 5.0 � 3.1 (UF) vs. 3.1 � 3.5kg (standard care); p ¼ 0.001

Dyspnea score: 5.4 � 1.1 (UF) vs.5.2 � 1.2 (standard care); p ¼ 0.588

90 days: HF rehospitalization:18% (UF) vs. 32% (standard care), p ¼ 0.022;

HR 0.56; 95% CI: 0.28–0.51; p ¼ 0.04Unscheduled clinic/emergency visits: 21% (UF)

vs. 44%, p ¼ 0.009

90 days: 9 (9.6%) UF vs.11 (11.6) standard care

No significant between-groupdifferences, except bleeding (1 UF vs.7 standard care, p ¼ 0.032).

UF group: 1 catheter infection, 5 filterclotting events, 1 patient transitionedto hemodialysis due to insufficientresponse to UF

Mean sCr change: þ0.23 � 0.70 mg/dl(UF) vs. �0.04 � 0.53 mg/dl (SPT)

Mean weight loss: 5.7 � 3.9 (UF) vs.5.5 � 5.1 kg (SPT); p ¼ 0.58

Crossover: STP: 6 patients STP: (6%) alsoreceived UF (2 before 96 h)

UF: 8 patients (9%) received diureticagents instead of UF; 28 patients(30%) also received diuretic agentsbefore 96 h.

7 days: no difference in death, worsening orpersistent HF, hemodialysis, SAE, orcrossover (23% UF vs. 18% SPT, p ¼ 0.45)

60 days HF hospitalization 26% (UF) vs. 26%(SPT) p ¼ 0.97

60 day: 17% UF vs.13% SPT; p ¼ 0.47

60-day SAE:72% UF vs. 57% SPT; p ¼ 0.03,

attributed to renal failure, bleeding,or catheter complications

3 (11%) UF vs. 14 (48%) standard care; HR:0.14; 95% CI: 0.04–0.48; p ¼ 0.002

Length of index hospitalization: 7.4� 4.6 (UF) vs.9.1� 1.9 days (standard care), p ¼ 0.23

Combined death or HF rehospitalizationHR for UF vs. standard care 0.35, 95% CI: 0.15–

0.69; p ¼ 0.0035

1 yr:7 (26%) UF vs. 11 (38%)

standard care; p ¼ 0.33

Premature clotting of filter in 6 patients

25% AUF vs. 35% ALD (p ¼ 0.11); HR:0.66; 95% CI: 0.4–1.1

Length of index hospitalization: median 6 (AUF)vs. 5 (ALD) days, p ¼ 0.106

30-day HF rehospitalizations/days at risk:11 of 2,876 (AUF) vs. 24 of 2,882 (ALD),p ¼ 0.06

30-day CV rehospitalizations/days at risk:17 of 2,882 (AUF) vs. 33 of 2,891 (ALD);p ¼ 0.037

For both HF and CV events: fewer patientsrehospitalized; fewer number of daysrehospitalized/days at risk

90 days 15% AUF vs. 13%ALD, p ¼ 0.83

At least 1 SAE: 66% (AUF) vs. 60%(ALD), p ¼ 0.4

SAEs of special interest: 23% (AUF) vs.14% (ALD); p ¼ 0.122

Related SAEs: 14.6% (UF) vs. 5.4%(ALD), p ¼ 0.026

Significant between group difference in% change from baseline in cardiacindex, CPO, dP/dtmax; no significantchange in sCr within or betweengroups

Signs/symptom score decreased significantly inboth groups; no difference between groups

Not reported Not reported


2435

not achieve statistical significance, several pre-specified secondary endpoints did. In particular, pa-tients in the adjustable UF group had significantlyfewer heart failure and cardiovascular events at 30days (56). Importantly, these events were adjudicatedby an independent committee blinded to randomizedtherapy (56). The finding of similar renal functionchanges in the 2 groups is consistent with that ofUNLOAD (33,56). In AVOID-HF, the average UF rate of138 ml/h was lower than the fixed 200-ml/h rate of theCARRESS-HF trial, and therapy was delivered over alonger period (70 vs. 41 h, respectively) (32,56).Adjustments of UF rates to individual patients’hemodynamics and renal function may explain thelack of differences in serum creatinine between

groups, despite a larger net fluid loss with UF(Figures 2 and 3) (56). Although they are detailed, thetherapy guidelines were adopted by the AVOID-HFinvestigators, most of whom continued to use themin their clinical practice. Individualization of fluidremoval rates may explain why the reduction in heartfailure events occurred earlier in AVOID-HF than inUNLOAD (33,56). Removal of isotonic fluid andavoidance of renin release by the macula densa maycontribute to the benefit of UF (32,54,55). Restorationof diuretic agent responsiveness may be a keymechanism by which UF delays recurrence of heartfailure events (56).

Significantly more patients in the UF group than inthe diuretic agents group experienced adverse events

FIGURE 2 Adjustable UF Guidelines Used by the AVOID-HF Investigators

Choose initial UF Rate

SBP 100-120 mm HgSBP < 100 mm Hg

150 cc/h

SBP > 120 mm Hg

250 cc/h200 cc/h

Decrease initial UF rate by 50 cc/hour if ANY of the following are present:

Every 6 hours, evaluate recent BP, HR, UO, net intake/output, sCr

• sCr rise >15% or 0.2 mg/dl compared to prior measurement• Resting SBP decreases >10 mm Hg compared to prior 6 h but remains >80 mm Hg• UO drops >50% compared to prior 6 h but remains >125 cc/6 h• Resting HR increase >20 bpm compared to prior 6 h but remains <120 bpm

• sCr rise >30% or 0.4 mg/dl compared to prior measurement• Resting SBP decreases >20 mm Hg compared to prior 6 h or >80 mm Hg• UO <125 cc/6 h• Resting HR increase >30 bpm compared to prior 6 h or >120 bpm

Consider decreasing UF rate by50 cc/hr and checking STAT sCr

If UF held, re-evaluate after laboratory values are available:

• If hemodynamics are stable and sCr has plateaued, then consider re-starting UF at rate 50-100 cc/hr less than previous rate

• If persistent, volume overload present, then consider • IV inotropes in patients with LVEF <40% or RV systolic dysfunction • Weaning vasodilators, especially in patients with HFpEF • RHC

Resolution of congestion (all of the following):• JVP <8 cm H2O• No orthopnea• Trace or no peripheral edema

None of these present Consider completion of UF therapy (See Figure 2, Panel B)

After completion of UF therapy

If satisfactory “dry weight” has beenreached AND sCr is stable:• Initiate oral loop diuretic therapy with goal to keep net even• GDMT

If sCr, hemodynamics or UO are NOTstable:• Hold diuretics until sCr is stable for a minimum of 12 h and then: • If “dry weight” /adequate decongestion has been reached then initiate oral diuretics with goal to keep net even • If “dry weight” /adequate decongestion has NOT been reached then initiate IV diuretics• If elevated sCr or hemodynamic instability present, then consider a bolus of IV fluid

Any of thesepresent

Persistent elevation insCr >1.0 mg/dl abovebaseline at start ofIV diuretic treatment

Best achievable “Dry Weight”• Evidence of poor tolerance of fluid removal -AND-• UF rate <100 cc/hr or Net negative < 1 liter/24 h

Persistenthemodynamicinstability

Strongly consider holding UF andchecking STAT sCr

• RV>LV dysfunction• sCr increase 0.3 mg/dl above recent baseline• Baseline sCr >2.0 mg/dl• History of instability with diuresis or UF in the past

A

B

(A) Guidelines for the adjustment of UF therapy. (B) Guidelines for the completion of ultrafiltration therapy: 40 mg of furosemide ¼ 1 mg

bumetanide or 10 mg of torsemide (52,53). b.i.d. ¼ twice daily; GDMT ¼ guideline-directed medical therapy; IV ¼ intravenous; JVP ¼ jugular

venous pressure; LV ¼ left ventricular; QD ¼ once daily; RV ¼ right ventricular; SBP ¼ systolic blood pressure; sCr ¼ serum creatinine;

UO ¼ urine output; other abbreviations as in Figure 1.



2436


2437

of special interest (infection requiring intravenousantibiotics, bleeding requiring transfusion, symp-tomatic hypotension requiring vasopressor agentsor rapid fluid replacement, a drop in hemoglobin>3 g/dl, and acute coronary syndrome requiringintervention [31% vs. 17%, respectively; p ¼ 0.018]).Serious therapy-related adverse events occurred athigher rates in the UF group than in the diureticagents group (14.6% vs. 5.4%, respectively; p ¼ 0.026)(56). Although in AVOID-HF, UF-related adverseevents were fewer than in CARRESS-HF, the excess oftherapy-related complications with UF is a seriousconcern (32,56). More study of the specifics ofproviding UF are needed to identify strategies aimedat minimizing access-related and other potentiallypreventable complications (32,56). Taken together,the facts outlined in the preceding text indicate thatthe AVOID-HF trial was unable to demonstrate thatadjustable UF is superior to adjustable diuretic agenttherapy. The statistically significant secondary out-comes of fewer 30-day heart failure and cardiovas-cular events are hypothesis-generating and do notdiminish the critical need for adequately poweredrandomized controlled trials comparing the effects ofUF versus diuretic agent-based strategies on bothheart failure morbidity and mortality.

KNOWLEDGE GAPS IN THE USE OF

EXTRACORPOREAL UF IN HEART FAILURE

SELECTION OF POTENTIAL CANDIDATES. The con-flicting results from UF studies highlight the fact thatpatient selection and fluid removal targets areincompletely understood (32,56). Practice guidelinessuggest that an inadequate response to an initialdose of an intravenous loop diuretic agent be treatedwith an increased dose of the same drug (54,55). Ifthis intervention is ineffective, invasive hemody-namic assessment is recommended. Evidence ofpersistent fluid excess can then be treated with theaddition of thiazide diuretic agents, aldosteroneantagonists, or continuous intravenous infusion of aloop diuretic agent. Only if all these measures failcan UF be considered (58,59). A similar degree ofdiuretic agent resistance characterized eligibility forenrollment in CARRESS-HF (32,33,53,56). In thistrial, the poor outcomes of UF in patients withthe acute cardiorenal syndrome may be partiallyrelated to the lack of therapy adjustment accordingto individual patients’ characteristics. In AVOID-HF,fine-tuning of UF rates in response to vital signs,renal function, or urine output resulted in greaternet fluid loss and was associated with fewer 30-dayheart failure events without a greater increase in

serum creatinine levels than in the adjustablediuretic agent group (56). These observations un-derscore the critical need for additional investigationof UF as both first-line and rescue therapies, pro-vided that UF rates are adjusted in each patient inresponse to changes in vital signs and renal function(32,56).

Due to the potential complications and cost of UF,it should not be used indiscriminately in decom-pensated heart failure. For example, in patients withde novo heart failure or those not receiving dailydiuretic agents, fluid overload can be rapidly elimi-nated with intravenous diuretic agents, which shouldbe used in such cases instead of UF. The unansweredquestion is which patients who develop heart failuredecompensation despite daily oral diuretic agentsshould be considered for UF instead of intravenousdiuretic agents? To date, all studies of UF in patientswith heart failure have relied solely on clinical signsand symptoms of fluid excess, both as inclusioncriteria and fluid removal targets. This is problematicdue to the poor correlation between clinical assess-ment and objective measures of increased fillingpressures (60). A European consensus statementthat graded congestion according to a combinationof clinical and laboratory parameters suggested thata score of $12, together with urine output of <1,000ml/24 h, should trigger the use of UF because thesevalues are indicative of diuretic agent resistance (61).This recommendation has not been prospectivelyvalidated and relies on the unproven assumption thatthe magnitude of fluid excess influences diureticagent responsiveness.

Data from 15 patients with acutely decompensatedheart failure show that urinary sodium concentrationin response to intravenous loop diuretic agents ishighly variable and lower than that in the ultrafiltrate(44). The difficulty in predicting the natriureticresponse of individual patients to a given dose ofintravenous diuretic agent is underscored by theabsence of a correlation between baseline renalfunction and urinary sodium concentration afterfurosemide administration (44). The hypothesis thatUF may be especially effective in patients withurinary sodium concentrations of <100 mEq after aspecified dose of intravenous diuretic agents shouldbe tested in randomized trials. A single non-randomized prospective cohort study showed similareffects of UF in heart failure with reduced versuspreserved left ventricular ejection fraction (62).However, because the 2 types of heart failure possessdistinct pathophysiological and clinical characteris-tics, response to UF should be assessed in controlledtrials.

FIGURE 3 Adjustable Loop Diuretic Agent Guidelines Used by the AVOID-HF Investigators

See table at bottom of figure†

Initial Diuretics Regimen

Continue currentdiuretic regimen

At 24 h If Persistent Volume Overload Present



Repeat 72 h Step Until Treatment Complete

≤8081-160161-240>240

+ or –+ or –+ or –+ or –

40 mg IV bolus + 5 mg/h80 mg IV bolus + 10 mg/h80 mg IV bolus + 20 mg/h80 mg bolus + 30 mg/h

05 mg metolazone QD5 mg metolazone BID5 mg metolazone BID

Loop (mg/day) Thiazide†Current dose Suggested dose

ThiazideLoop

UO 3- 5 L/dayUO > 5 L/day

Reduce currentdiuretic regimenif desired*

UO < 3 L/day

Advance to next step on table†Continue currentdiuretic regimen



UO < 3 L/day




UO < 3 L/day




UO < 3 L/day

Advance to next step on table†

and consider:a. IV inotropes if SBP <110 mm Hgand LVEF <40% or RVsystolic dysfunctionb. NTG or nesiritide ifSBP >120 mm Hg (any LVEF) andsevere symptoms

Advance to next step on table†

and consider:a. IV inotropes if SBP <110 mm Hgand LVEF <40% or RVsystolic dysfunctionb. NTG or nesiritide ifSBP >120 mm Hg (any LVEF) andsevere symptomsc. Right heart catheterization

A

(A) Initiation of loop diuretic agents. *Evaluation of blood pressure, heart rate, urine output, and net intake/output was performed every 6 h; evaluation

of serum chemistries was performed every 12 h. Decreasing or holding the diuretic agent dose may be considered if: 1) serum creatinine rises by 30%

or $0.4 mg/dl (whichever is less) versus previous measurement; 2) resting systolic blood pressure decreases >20 mm Hg compared to previous 6 h

or drops <80 mm Hg; or 3) resting heart rate is >30 beats/min compared to previous 6 h or >120 beats/min. LVEF ¼ left ventricular ejection fraction;

NTG ¼ nitroglycerin; other abbreviations as in Figure 2.

Continued on the next page



2438

FIGURE 3 Continued

Consider Completion of Therapy if ONE of the following:

Resolution of congestion(all of the following):• JVP <8 cm H2O• No orthopnea• Trace or no peripheral edema

Best achievable “dry weight”has been achieved• Hemodynamic evidence of poortolerance of fluid removal bypersistent hemodynamic changes-AND-• Net negative <1 l/24 h

Persistent elevation insCr >1.0 mg/dL above baselineat start of IV diuretic treatment

If satisfactory “dry weight” has been reachedAND sCr is stable• Initiate loop diuretic therapy with goal to keep net even• GDMT

If sCr, hemodynamics or UO are NOT stable• Hold diuretics until sCr is stable for a minimum of 12 h and then initiate oral diuretics as above• If elevated sCr or hemodynamic instability present, then consider a bolus of IV fluid

After Completion of IV Loop Diuretic Therapy

Persistent hemodynamicinstability

B

C

(B) Guidelines for the completion of adjustable loop diuretic agents. (C) Guidelines for management after completion of adjustable loop diuretic agents

(see also references 52 and 53).


2439

FLUID REMOVAL TARGETS AND MONITORING OF UF

THERAPY. One important general recommendation isthat, once an initial UF rate has been chosen, it shouldbe either maintained or reduced because capillaryrefill from the interstitium decreases as fluid isremoved (34). Although the optimal rate and durationof UF must be individualized, UF rates of >250 ml/hare not typically recommended (56,57). Patients withpredominantly right-sided heart failure or patientswith heart failure with preserved ejection fraction areexquisitely susceptible to intravascular volumedepletion and may only tolerate low UF rates (50 to100 ml/h) (63). In addition, clinical experienceteaches that extracorporeal fluid removal is bettertolerated when conducted with low UF rates overprolonged periods of time (32).

A frequently used approach is to compare patients’current weight with that preceding the signs andsymptoms of congestion and to use this “dry weight”as the target for fluid removal. No consensus exists onwhether removal of only 60% to 80% of excess fluid byUF and continuation of loop diuretic agents duringtherapy results in less hemodynamic instability andgreater urinary sodium excretion (53). Considering theharmful renal effects of an increased central venous

pressure (9,13–15,34,35,64), controlled clinical trialsshould determine if fluid removal by UF should beadjusted to achieve specific central venous pressuretargets. In lieu of invasive measurements, ultraso-nography can help estimate central venous pressurewith the assessment of the respiratory excursions ofthe diameter of the inferior vena cava (65). Althoughultrasonography is noninvasive and inexpensive, itsreliability depends strictly depends on the operator’sskill and the patient’s respiratory effort (65).

Studies of implantable hemodynamic monitorshave consistently shown that baseline pulmonaryartery diastolic pressure predicts heart failure events.Interventions aimed at reducing pulmonary arterypressures to pre-specified target ranges have effec-tively reduced heart failure events without significantrenal function changes (66,67). The CardioMEMSsensor (St. Jude Medical, St. Paul, Minnesota) permitsmeasurement of pulmonary artery pressures asfrequently as clinically indicated. Therefore, it isconceivable that, in patients in whom theCardioMEMS device has been implanted, fluid can beremoved by UF until the target range of pulmonaryartery pressures that effectively reduced heart failureevents has been achieved (66,67).



2440

BLOOD VOLUME AND FLUID EXCESS ESTIMATION.

The hematocrit is the ratio of the volume occupied byred blood cells to that of whole blood. Because redblood cell mass does not change in the short termunless bleeding occurs, fluctuations in hematocritreflect changes in intravascular volume (68).

Online hematocrit sensors permit continuous esti-mation of blood volume changes during UF and canbe programmed to stop fluid removal if the hemato-crit exceeds a threshold set by the clinician (e.g., 5%to 7%) and resume therapy when the hematocritvalue falls below the pre-specified limit, indicating anadequate refilling of the intravascular volume fromthe interstitial space (Online Figure 1) (68).

However, because numerous factors (e.g., changein body position) can alter hematocrit values,physical, laboratory, and hemodynamic variablesshould be concomitantly assessed to determine theappropriate UF rates and the amount of fluid thatshould be removed (68). Bioimpedance vectoranalysis relies on the principle that whole-bodyimpedance to an alternating current reflects totalbody water (r ¼ 0.996) (69). Measurements of bio-impedance vector require 2 pairs of electrodes beplaced on the wrist and ankles and the application ofa 50-kHz alternating microcurrent (CardioEFG, EFGDiagnostics, Belfast, Northern Ireland) (69). It istherefore attractive to envision the use of bio-impedance vector analysis to determine baselinefluid status and then use of serial measurements toguide the amount and rate of fluid removal by UF ordiuretic agents. Accuracy of bioimpedance vectoranalysis can be reduced by diaphoresis, hirsutism,incorrect electrode placement, cutaneous alterations,or improper electrical grounding. Bioimpedancespectroscopy is also being investigated in patientswith heart failure (70). Unfortunately, no existingbioimpedance-based method can differentiate intra-vascular from interstitial extracellular fluid volume, adistinction that is critical for safe and effective fluidremoval (69,70).

Intrathoracic fluid can also be measured non-invasively with electromagnetic technology insertedin a removable vest (Sensible Medical Innovations,Netanya, Israel). This device, shown in preclinical andpilot human studies to measure intrathoracic water asaccurately as computed tomography, is being testedin a prospective, randomized clinical trial(NCT02448342) (71). The measurement of blood vol-ume using iodine-131-labeled albumin is accurate, butthe 6 to 9 blood draws needed to create the dilutioncurve make it impractical for the serial assessmentsneeded during fluid removal (72). The lack of optimalmethods for the estimation of blood volume and fluid

excess underscores the critical need for research inthis area.

BIOMARKERS. The use of natriuretic peptides toassess volume status and guide decongestive thera-pies cannot be recommended because fluid overload isnot the sole cause of increases in the levels of thesebiomarkers (73). The removal of fluid to achievepre-specified natriuretic peptide levels is untested inacute heart failure. Serum creatinine is the solebiomarker used to guide fluid removal because of thebelief that its level reflects both renal filtration func-tion and tubular status. However, serum creatininewas established and validated as a measurement ofrenal function only at the point of steady-state(constant production from the metabolism of musclecreatine phosphate and unchanging glomerular filtra-tion and urinary flow to excrete creatinine at a con-stant rate). Therefore, it is unfortunate that serumcreatinine is the only widely available measurement ofrenal function in patients with acute illnesses, such asacutely decompensated heart failure, where the ratesof creatinine production and excretion may be altered.Gene expression analysis has shown differences in thegenes expressed in acute kidney injury due to differentprocesses, even if the magnitude of rise in creatinine isthe same. Conversely, serum creatinine concentrationcan be normal with documented tubular injury due tothe delayed achievement of detectable changes of thisanalyte (74). Generally, hemodynamically drivenincreases in serum creatinine resolve with treatmentin 24 to 72 h, whereas the cellular derangements due toacute tubular damage, or even necrosis, may last forweeks (75). Therefore, the duration of the elevation inserum creatinine has a greater predictive effect onmorbidity and mortality than the extent of this bio-marker’s elevation (76,77). Indeed, the use of increasesin serum creatinine as an endpoint for acutelydecompensated heart failure trials has been chal-lenged. Evaluation of the relationship betweenchanges in serum creatinine and 60-day outcomes inDOSE subjects revealed that increases in serum creat-inine from baseline to 72 h (DOSE’s coprimaryendpoint) was associated with lower risk for the com-posite outcome of death or heart failure events.Conversely, there was a strong relationship betweenimproved renal function and unfavorable 60-dayoutcomes (78). Thus, serum creatinine changes arean unreliable surrogate endpoint in trials of fluidremoval therapies. After the discovery by Mishra et al.(79) of neutrophil gelatinase-associated lipocalin(NGAL), which is secreted in the urine and the plasmaby a damaged kidney, it was shown that the expres-sion/secretion of urine NGAL (neutrophil gelatinase-


https://clinicaltrials.gov/ct2/show/NCT02448342?term=NCT02448342&rank=1


2441

associated lipocalin) occurred within 3 h of the event(sepsis, nephrotoxins, obstruction, ischemia); andthat the amount of secreted protein (from 20 ng/ml to5 mg/ml) was proportional to the severity and time ofresolution of the stimulus. A growing body of evidencesuggests NGAL is not expressed when serum creati-nine increases due to volume stressors. A systematicstudy of thousands of genes encoding for several bio-markers including NGAL, kidney injury molecule-1,tissue inhibitor of metalloproteinase-1, and clusterin,found that these molecules were detectable after abrief dose of ischemia, yet none of these genes wereexpressed after near-fatal volume depletion, despitethe rise in serum creatinine in both models (80).Although thismethod is not yetwidely available, in thesetting of any method of fluid removal, the levels ofurine NGAL and other biomarkers of tubular injurycould potentially help distinguish a rise in serumcreatinine due to a hemodynamically mediateddecrease in glomerular filtration rate or actual tubularinjury (74). Numerous genes are differentiallyexpressed depending upon the presence and type ofacute kidney injury. The levels of NGAL rise faster thanthose of other indicators of renal injury, makingthis biomarker better suited to distinguish betweenhemodynamically, versus tubular injury-drivenincreases in serum creatinine that may occur duringfluid removal therapies (74,80).

ECONOMIC CONSIDERATIONS. To date, there hasbeen no prospective evaluation of the cost effective-ness of UF therapy. The only published retrospectiveestimate of UF costs is presented in the OnlineAppendix in Section 3.0. Therefore, it is imperativethat future prospective controlled trials includerigorous cost-benefit analyses.

PROPOSAL FOR FUTURE STUDIES. No studies per-formed to date have conclusively demonstrated thesuperiority of one fluid removal method over another.It is vital to continue searching for the most effectiveand safest method to treat congestion, which worsensthe outcomes of patients with heart failure and cau-ses unacceptably high hospitalization rates world-wide. Concerning UF, priority should be given tomechanistic studies including evaluation of diureticagent responsiveness at baseline during and afterfluid removal by using the measurements describedin this review (21). Hemodynamic measurements thatreflect fluid status (e.g., central venous pressure andpulmonary artery diastolic pressure) should also beperformed at baseline and throughout therapy. Spe-cific hemodynamic targets indicative of optimal fluidstatus should be established in individual patients,similar to the strategies used to guide medication

adjustment in studies of pulmonary artery pressuresensors (63). Different UF rates should be tested interms of their ability to reach these hemodynamictargets without causing renal tubular damage, asdetectable by increase in urine levels of biomarkers,such as NGAL (70,73,76). This will require simulta-neous measurement of the selected hemodynamicvalues and biomarker levels capable of differentiatingrises in serum creatinine due to decreases inglomerular filtration rate produced by intravascularfluid removal from those reflective of renal injury.Serial measurements of urine and ultrafiltrate sodiumcontent (rather than randomly performed single-spotmeasurements) may also help to better characterizeand compare the amount and pattern of sodiumextraction during UF therapy and conventionaldiuretic agent-based regimens. This noninvasive,inexpensive, and readily available test can easily beincorporated into future investigations. The results ofmechanistic studies are essential to determine howfluid removal rates and amounts should be adjustedin individual subjects of future controlled trials(“precision” fluid removal).

Equally important is the development of vascularaccesses and UF device components that increase theefficiency and safety of the therapy. The device- andtherapy-related adverse events observed in previoustrials should undergo careful re-evaluation to deter-mine which were preventable or related to operatorexperience versus those that were inherent to howtherapy was delivered or was unpredictable(32,46,47,52).

Only after these issues have been satisfactorilyaddressed should a carefully designed, adequatelypowered study be considered to prospectivelycompare UF with pharmacological fluid removaltherapies. All treatments should be tailored to indi-vidual patients’ hemodynamic and renal status. Inaddition, the study’s follow-up period should be suf-ficiently long to permit the evaluation of morbidity(rehospitalizations) and mortality. Future trialsshould also evaluate whether the greater cost of me-chanical fluid removal during the index hospitaliza-tion is offset by the savings resulting from potentiallyfewer heart failure events in patients treated with UF.

As the cost of inpatient care is very high, seriousconsideration should be given to studies in theoutpatient setting to determine the relative safetyand effectiveness of intermittent pharmacologicaland mechanical fluid removal therapies for the pre-vention rather than the treatment of heart failurehospitalizations. Intermittent outpatient UF torestore responsiveness to oral diuretic agents is also astrategy that deserves investigation. Finally,



CENTRAL ILLUSTRATION Ultrafiltration for Fluid Overload in Heart Failure

Decreased cardiac output due to chronic heart failure

Cardio-renal syndrome:Abnormal hemodynamics, neurohormonal activation, excessive tubular sodium reabsorption,

Decreased water clearance and increased sodium reabsorption

Unpredictable elimination of sodium and water

Development of diuretic resistance

Risk of hypokalemia (low potassium levels)and hypomagnesemia (low magnesium levels)

Persistent congestion, failure to lower sodium levels

Worsening heart failure, increased mortalityafter discharge, increase in re-hospitalization rates

LOOP DIURETICSto eliminate hypotonic urine

Restoration of diuretic responsiveness

No change in electrolytes, particularly potassium and magnesium

heart failure events compared to loop diuretics

and improved outcomes

ULTRAFILTRATIONto remove isotonic plasma water

Costanzo, M.R. et al. J Am Coll Cardiol. 2017;69(19):2428–45.

Of the >1 million heart failure hospitalizations in the United States and Europe, 90% are due to signs and symptoms of fluid overload. This enormous worldwide health

care burden is aggravated by the fact that recurrent congestion worsens patients’ outcomes, regardless of age and renal function. Abnormal hemodynamics,

neurohormonal activation, excessive tubular sodium reabsorption, inflammation, oxidative stress, and nephrotoxic medications drive the complex interactions between

heart and kidney (cardiorenal syndrome). Loop diuretic agents are used in most congested patients. Due to their mechanism and site of action, loop diuretic agents

lead to the production of hypotonic urine and may contribute to diuretic agent resistance (“braking phenomenon,” distal tubular adaptation, and increased renin

secretion in the macula densa). Increased uremic anions and proteinuria also impair achievement of therapeutic concentrations at their tubular site of action.

Ultrafiltration is the production of plasma water from whole blood across a hemofilter in response to a transmembrane pressure. Therefore, ultrafiltration removes

isotonic fluid without direct activation of the renin-angiotensin-aldosterone system, provided that fluid removal rates do not exceed capillary refill. Any method of

fluid removal may cause an increase in serum creatinine. However, in the absence of evidence of renal tubular injury (e.g., augmented urinary concentration of

neutrophil gelatinase-associated lipocalin), this increase represents a physiological decrease in glomerular filtration rate due to decreased intravascular volume from

fluid removal. AVP ¼ arginine vasopressin; GFR ¼ glomerular filtration rate; K ¼ potassium; KIM ¼ kidney injury molecule; Mg ¼ magnesium; NGAL ¼ neutrophil

gelatinase-associated lipocalin; RAAS ¼ renin-angiotensin-aldosterone system; SNS ¼ sympathetic nervous system.



2442

technological advances may permit the developmentof “wearable” UF devices capable of delivering indi-vidualized UF therapy.

CONCLUSIONS

Fluid excess drives most heart failure hospitaliza-tions. Recurrent hospitalizations are common andpredict unfavorable outcomes. As heart failure pro-gresses, a significant proportion of patients develop

an inadequate response to diuretic agent therapy.Additional approaches, such as sequential nephronalblockade with thiazide diuretic agents or high-dosealdosterone antagonists, have not been appropri-ately validated. Other pharmacological therapieshave not improved the outcomes of patients withheart failure who have fluid overload. Ultrafiltrationis an attractive alternative therapy because it pre-dictably removes total body sodium. In futurestudies, UF should be adjusted according to the


2443

patient’s hemodynamic and renal profiles; andpatient selection, fluid removal amount, duration,and rate should be guided by objective, complemen-tary, and informative measurements of fluid overloadand kidney function (Central Illustration). Theurgency of these investigations is underscored by thealarming prognostic and economic implications ofrecurrent heart failure hospitalizations, which remainunacceptably high with conventional pharmacolog-ical therapies.

ACKNOWLEDGMENT The authors thank Wendy Gat-tis Stough, PharmD, who worked under the directionand supervision of Dr. Costanzo, for editing assistance.

ADDRESS FOR CORRESPONDENCE: Dr. Maria RosaCostanzo, Advocate Medical Group Midwest HeartSpecialists, Edward Heart Hospital, 4th Floor, 801South Washington Street, PO Box 3226, Naperville,Illinois 60566. E-mail: [email protected].

RE F E RENCE S

1. Ambrosy AP, Fonarow GC, Butler J, et al. Theglobal health and economic burden of hospitali-zations for heart failure: lessons learned fromhospitalized heart failure registries. J Am CollCardiol 2014;63:1123–33.

2. Crespo-Leiro MG, Anker SD, Maggioni AP, et al.,Heart Failure Association (HFA) of the EuropeanSociety of Cardiology (ESC). European Society ofCardiology Heart Failure Long-Term Registry(ESC-HF-LT): 1-year follow-up outcomes anddifferences across regions. Eur J Heart Fail 2016;18:613–25.

3. Setoguchi S, Stevenson LW, Schneeweiss S.Repeated hospitalizations predict mortality in thecommunity population with heart failure. AmHeart J 2007;154:260–6.

4. McKie PM, Schirger JA, Costello-Boerrigter LC,et al. Impaired natriuretic and renal endocrineresponse to acute volume expansion in pre-clinicalsystolic and diastolic dysfunction. J Am Coll Car-diol 2011;58:2095–103.

5. Peacock WF IV, De Marco T, Fonarow GC, et al.,for the ADHERE investigators. Cardiac troponinand outcome in acute heart failure. N Engl J Med2008;358:2117–26.

6. Rubboli A, Sobotka PA, Euler DE. Effect ofacute edema on left ventricular function andcoronary vascular resistance in the isolated ratheart. Am J Physiol 1994;267:H1054–61.

7. Verbrugge FH, Bertrand PB, Willems E, et al.Global myocardial oedema in advanced decom-pensated heart failure. Eur Heart J CardiovascImaging 2016 Jul 4 [E-pub ahead of print].

8. Verbrugge FH, Dupont M, Steels P, et al. Thekidney in congestive heart failure: “are natri-uresis, sodium, and diuretics really the good, thebad and the ugly?”. Eur J Heart Fail 2014;16:133–42.

9. Braam B, Cupples WA, Joles JA, Gaillard C.Systemic arterial and venous determinants of renalhemodynamics in congestive heart failure. HeartFail Rev 2012;17:161–75.

10. Ronco C, Haapio M, House AA, Anvekar N,Bellomo R. Cardiorenal syndrome. J Am CollCardiol 2008;52:1527–39.

11. Firth JD, Raine AE, Ledingham JG. Raisedvenous pressure: a direct cause of renal sodiumretention in oedema? Lancet 1988;1:1033–5.

12. Winton FR. The influence of venous pressureon the isolated mammalian kidney. J Physiol 1931;72:49–61.

13. Damman K, van Deursen VM, Navis G,Voors AA, van Veldhuisen DJ, Hillege HL.Increased central venous pressure is associatedwith impaired renal function and mortality in abroad spectrum of patients with cardiovasculardisease. J Am Coll Cardiol 2009;53:582–8.

14. Mullens W, Abrahams Z, Francis GS, et al.Importance of venous congestion for worsening ofrenal function in advanced decompensated heartfailure. J Am Coll Cardiol 2009;53:589–96.

15. Colombo PC, Onat D, Harxhi A, et al. Peripheralvenous congestion causes inflammation, neuro-hormonal, and endothelial cell activation. EurHeart J 2014;35:448–54.

16. Metra M, Davison B, Bettari L, et al. Is wors-ening renal function an ominous prognostic sign inpatients with acute heart failure? The role ofcongestion and its interaction with renal function.Circ Heart Fail 2012;5:54–62.

17. Testani JM, Chen J, McCauley BD, Kimmel SE,Shannon RP. Potential effects of aggressivedecongestion during the treatment of decom-pensated heart failure on renal function andsurvival. Circulation 2010;122:265–72.

18. Damman K, Ng Kam Chuen MJ, MacFadyen RJ,et al. Volume status and diuretic therapy in sys-tolic heart failure and the detection of early ab-normalities in renal and tubular function. J Am CollCardiol 2011;57:2233–41.

19. Thomas ME, Blaine C, Dawnay A, et al. Thedefinition of acute kidney injury and its use inpractice. Kidney Int 2015;87:62–73.

20. Singh D, Shrestha K, Testani JM, et al. Insuf-ficient natriuretic response to continuous intrave-nous furosemide is associated with poor long-termoutcomes in acute decompensated heart failure.J Card Fail 2014;20:392–9.

21. ter Maaten JM, Valente MA, Damman K,Hillege HL, Navis F, Voors AA. Diuretic response inacute heart failure—pathophysiology, evaluation,and therapy. Nat Rev Cardiol 2015;12:184–92.

22. Voors AA, Davison BA, Teerlink JR, et al., forthe RELAX-AHF investigators. Diuretic response inpatients with acute decompensated heart failure:characteristics and clinical outcome—an analysis

from RELAX-AHF. Eur J Heart Fail 2014;16:1230–40.

23. Gheorghiade M, Filippatos G. Reassessingtreatment of acute heart failure syndromes: theADHERE registry. Eur Heart J Suppl 2005;7:B13–9.

24. Felker GM, Lee KL, Bull DA, et al., for theNHLBI Heart Failure Clinical Research Network.Diuretic strategies in patients with acute decom-pensated heart failure. N Engl J Med 2011;364:797–805.

25. Konstam MA, Gheorghiade M, Burnett JC Jr.,et al., for the Efficacy of Vasopressin Antagonismin Heart Failure Outcome Study With Tolvaptan(EVEREST) investigators. Effects of oral tolvaptanin patients hospitalized for worsening heart fail-ure: the EVEREST outcome trial. JAMA 2007;297:1319–31.

26. Massie BM, O’Connor CM, Metra M, et al., forthe PROTECT investigators and committees.Rolofylline, an adenosine A1-receptor antagonist,in acute heart failure. N Engl J Med 2010;363:1419–28.

27. O’Connor CM, Starling RC, Hernandez AF, et al.Effect of nesiritide in patients with acute decom-pensated heart failure [published correctionappears in N Engl J Med 2011;365:773]. N Engl JMed 2011;365:32–43.

28. Costanzo MR, Jessup M. Treatment ofcongestion in heart failure with diuretics andextracorporeal therapies: effects on symptoms,renal function, and prognosis. Heart Fail Rev 2012;17:313–24.

29. Ronco C, Ricci Z, Bellomo R, Bedogni F.Extracorporeal ultrafiltration for the treatment ofoverhydration and congestive heart failure.Cardiology 2001;96:155–68.

30. Jaski BE, Ha J, Denys BG, Lamba S, Trupp RJ,Abraham WT. Peripherally inserted veno-venousultrafiltration for rapid treatment of volumeoverloaded patients. J Card Fail 2003;9:227–31.

31. Agostoni P, Marenzi G, Lauri G, et al. Sustainedimprovement in functional capacity after removalof body fluid with isolated ultrafiltration inchronic cardiac insufficiency: failure of furosemideto provide the same result. Am J Med 1994;96:191–9.

32. Bart BA, Goldsmith SR, Lee KL, et al., for theHeart Failure Clinical Research Network. Ultrafil-tration in decompensated heart failure with

mailto:[email protected]

mailto:[email protected]

http://refhub.elsevier.com/S0735-1097(17)36697-4/sref1



















































































































































2444

cardiorenal syndrome. N Engl J Med 2012;367:2296–304.

33. Costanzo MR, Guglin ME, Saltzberg MT, et al.,for the UNLOAD trial investigators. Ultrafiltrationversus intravenous diuretics for patients hospital-ized for acute decompensated heart failure. J AmColl Cardiol 2007;49:675–83.

34. Marenzi G, Grazi S, Giraldi F, et al. Interrelationof humoral factors, hemodynamics, and fluid andsalt metabolism in congestive heart failure: effectsof extracorporeal ultrafiltration. Am J Med 1993;94:49–56.

35. Marenzi G, Lauri G, Grazi M, Assanelli E,Campodonico J, Agostoni P. Circulatory responseto fluid overload removal by extracorporealultrafiltration in refractory congestive heart fail-ure. J Am Coll Cardiol 2001;38:963–8.

36. Agostoni PG, Marenzi GC, Pepi M, et al. Iso-lated ultrafiltration in moderate congestive heartfailure. J Am Coll Cardiol 1993;21:424–31.

37. Donato L, Biagini A, Contini C, et al. Treatmentof end-stage congestive heart failure by extra-corporeal ultrafiltration. Am J Cardiol 1987;59:379–80.

38. Akiba T, Taniguchi K, Marumo F, Matsuda O.Clinical significance of renal hemodynamics in se-vere congestive heart failure: responsiveness toultrafiltration therapies. Jpn Circ J 1989;53:191–6.

39. Costanzo MR, Saltzberg M, O’Sullivan J,Sobotka P. Early ultrafiltration in patients withdecompensated heart failure and diuretic resis-tance. J Am Coll Cardiol 2005;46:2047–51.

40. Bart BA, Boyle A, Bank AJ, et al. Ultrafiltrationversus usual care for hospitalized patients withheart failure: the Relief for Acutely Fluid-Overloaded Patients With DecompensatedCongestive Heart Failure (RAPID-CHF) trial. J AmColl Cardiol 2005;46:2043–6.

41. Rogers HL, Marshall J, Bock J, et al.A randomized, controlled trial of the renal effectsof ultrafiltration as compared to furosemide inpatients with acute decompensated heart failure.J Card Fail 2008;14:1–5.

42. Elkayam U, Hatamizadeh P, Janmohamed M.The challenge of correcting volume overload inhospitalized patients with decompensated heartfailure. J Am Coll Cardiol 2007;49:684–6.

43. Costanzo MR, Saltzberg MT, Jessup M,Teerlink JR, Sobotka PA, for the UltrafiltrationVersus Intravenous Diuretics for Patients Hospi-talized for Acute Decompensated Heart Failure(UNLOAD) investigators. Ultrafiltration is associ-ated with fewer rehospitalizations than continuousdiuretic infusion in patients with decompensatedheart failure: results from UNLOAD. J Cardiac Fail2010;16:277–84.

44. Ali SS, Olinger CC, Sobotka PA, et al. Loopdiuretics can cause clinical natriuretic failure: aprescription for volume expansion. Congest HeartFail 2009;15:1–4.

45. Giglioli C, Landi D, Cecchi E, et al. Effectsof ULTRAfiltration vs. DIureticS on clinical,biohumoral and haemodynamic variables in pa-tients with deCOmpensated heart failure: theULTRADISCO study. Eur J Heart Fail 2011;13:337–46.

46. Agostoni PG, Marenzi GC, Sganzerla P, et al.Lung-heart interaction as a substrate for theimprovement in exercise capacity after body fluidvolume depletion in moderate congestive heartfailure. Am J Cardiol 1995;76:793–8.

47. Pepi M, Marenzi GC, Agostoni PG, et al. Sus-tained cardiac diastolic changes elicited by ultra-filtration in patients with moderate congestiveheart failure: pathophysiological correlates. BrHeart J 1993;70:135–40.

48. Bart BA, Goldsmith SR, Lee KL, et al. Car-diorenal rescue study in acute decompensatedheart failure: rationale and design of CARRESS-HF,for the Heart Failure Clinical Research Network.J Card Fail 2012;18:176–82.

49. Mentz RJ, Stevens SR, DeVore AD, et al.Decongestion strategies and renin-angiotensin-aldosterone system activation in acute heart fail-ure. J Am Coll Cardiol HF 2015;3:97–107.

50. Dev S, Shirolkar SC, Stevens SR, et al. Reduc-tion in body weight but worsening renal functionwith late ultrafiltration for treatment of acutedecompensated heart failure. Cardiology 2012;123:145–53.

51. Patarroyo M, Wehbe E, Hanna M, et al. Car-diorenal outcomes after slow continuous ultrafil-tration therapy in refractory patients withadvanced decompensated heart failure. J Am CollCardiol 2012;60:1906–12.

52. Tang WH. Reconsidering ultrafiltration in theacute cardiorenal syndrome. N Engl J Med 2012;367:2351–2.

53. Marenzi G, Muratori M, Cosentino ER, et al.Continuous ultrafiltration for congestive heartfailure: the CUORE trial. J Card Fail 2014;20:9–17.

54. Lorenz JN, Weihprecht H, Schnermann J,Skøtt O, Briggs JP. Renin release from isolatedjuxtaglomerular apparatus depends on maculadensa chloride transport. Am J Physiol 1991;260:F486–93.

55. Schlatter E, Salomonsson M, Persson AE,Greger R. Macula densa cells sense luminal NaClconcentration via furosemide sensitive Naþ2Cl-Kþcotransport. Pflugers Arch 1989;414:286–90.

56. Costanzo MR, Negoianu D, Jaski BE, et al.Aquapheresis versus intravenous diuretics andhospitalizations for heart failure. J Am Coll CardiolHF 2016;4:95–105.

57. Costanzo MR, Negoianu D, Fonarow GC, et al.Rationale and design of the Aquapheresis VersusIntravenous Diuretics and Hospitalization for HeartFailure (AVOID-HF) trial. Am Heart J 2015;170:471–82.

58. Ponikowski P, Voors AA, Anker SD, et al. 2016ESC Guidelines for the diagnosis and treatment ofacute and chronic heart failure: the task force forthe diagnosis and treatment of acute and chronicheart failure of the European Society of Cardiology(ESC). Developed with the special contribution ofthe Heart Failure Association (HFA) of the ESC. EurHeart J 2016;37:2129–200.

59. Yancy CW, Jessup M, Bozkurt B, et al. 2013ACCF/AHA guideline for the management of heartfailure: a report of the American College of Car-diology Foundation/American Heart Association

task force on practice guidelines. J Am Coll Cardiol2013;62:e147–239.

60. Stevenson LW, Perloff JK. The limited reli-ability of physical signs for estimating hemody-namics in chronic heart failure. JAMA 1989;261:884–8.

61. Gheorghiade M, Follath F, Ponikowski P, et al.Assessing and grading congestion in acute heartfailure: a scientific statement from the acute heartfailure committee of the heart failure associationof the European Society of Cardiology andendorsed by the European Society of IntensiveCare Medicine. Eur J Heart Fail 2010;12:423–33.

62. Jefferies JL, Bartone C, Menon S, Egnaczyk GF,O’Brien TM, Chung ES. Ultrafiltration in heartfailure with preserved ejection fraction: compari-son with systolic heart failure patients. Circ HeartFail 2013;6:733–9.

63. Schrier RW, Bansal S. Pulmonary hypertension,right ventricular failure, and kidney: different fromleft ventricular failure? Clin J Am Soc Nephrol2008;3:1232–7.

64. Ross EA. Congestive renal failure: the patho-physiology and treatment of renal venous hyper-tension. J Card Fail 2012;18:930–8.

65. Stawicki SP, Braslow BM, Panebianco NL, et al.Intensivist use of hand-carried ultrasonography tomeasure IVC collapsibility in estimating intravas-cular volume status: correlations with CVP. J AmColl Surg 2009;209:55–61.

66. Abraham WT, Adamson PB, Bourge RC, et al.,for the CHAMPION trial study group. Wirelesspulmonary artery haemodynamic monitoring inchronic heart failure: a randomised controlled trial[published correction appears in Lancet 2012;379:412]. Lancet 2011;377:658–66.

67. Costanzo MR, Stevenson LW, Adamson PB,et al. Interventions linked to decreased heartfailure hospitalizations during ambulatory pulmo-nary artery pressure monitoring. J Am Coll CardiolHF 2016;4:333–44.

68. Ronco C, Brendolan A, Bellomo R. Onlinemonitoring in continuous renal replacement ther-apies. Kidney Int Suppl 1999;72:S8–14.

69. Piccoli A. Whole body-single frequency bio-impedance. Contrib Nephrol 2005;149:150–61.

70. Ribas N, Nescolarde L, Domingo M,Gastelurrutia P, Bayés-Genis A, Rosell-Ferrer J.Longitudinal and transversal bioimpedance mea-surements in addition to diagnosis of heart failure.J Phys Conf Ser 2010;224:012099.

71. Abraham WT, Amir O, Weinstein JM, Abbo A,Ben Gal T. Remote dielectric sensing (ReDS)-guided patient management of ambulatory heartfailure patients reduces rehospitalization rates[abstr]. J Card Fail 2015;21:S77.

72. Margouleff D. Blood volume determination, anuclear medicine test in evolution. Clin Nucl Med2013;38:534–7.

73. Bayes-Genis A, Lupón J, Jaffe AS. Can natri-uretic peptides be used to guide therapy? EJIFCC2016;27:208–16.

74. Sise ME, Forster C, Singer E, et al. Urineneutrophil gelatinase-associated lipocalin iden-tifies unilateral and bilateral urinary tract







































































































































































































2445

obstruction. Nephrol Dial Transplant 2011;26:4132–5.

75. Brown JR, Kramer RS, Coca SG, Parikh CR.Duration of acute kidney injury impacts long-termsurvival after cardiac surgery. Ann Thorac Surg2010;90:1142–8.

76. Parikh CR, Coca SG. Acute kidney injury:defining prerenal azotemia in clinical practice andresearch. Nat Rev Nephrol 2010;6:641–2.

77. Nickolas TL, O’Rourke MJ, Yang J, et al.Sensitivity and specificity of a single emergencydepartment measurement of urinary neutrophilgelatinase-associated lipocalin for diagnosing

acute kidney injury. Ann Intern Med 2008;148:810–9.

78. Brisco MA, Zile MR, Hanberg JS, et al. Rele-vance of changes in serum creatinine during aheart failure trial of decongestive strategies: in-sights from the DOSE trial. J Card Fail 2016;22:753–60.

79. Mishra J, Ma Q, Prada A, et al. Identification ofneutrophil gelatinase-associated lipocalin as anovel early urinary biomarker for ischemic renalinjury. J Am Soc Nephrol 2003;14:2534–43.

80. Xu K, Rosenstiel P, Paragas N, et al. Uniquetranscriptional programs identify subtypes of AKI.

J Am Soc Nephrol 2016 Dec 27 [E-pub ahead ofprint].

KEY WORDS biomarkers, creatinine,diuretics, glomerular filtration rate, venouscongestion

APPENDIX For an expanded Methodssection as well as supplemental tables and afigure, please see the online version of thisarticle.





























extracorporeal ultraﬁltration for fluidoverloadinheartfailure · dan negoianu, md,q margaret m....

Documents